GPU comparison
MacBook Pro M5 Max (36GB) vs NVIDIA RTX 5090 (32GB) for local LLMs
How the two stack up for running open-source LLMs locally — memory, bandwidth, price, and how many of the 48 tracked models each one can run.
| MacBook Pro M5 Max (36GB) | NVIDIA RTX 5090 (32GB) | |
|---|---|---|
| Memory | 36.0 GB | 32.0 GB |
| Bandwidth | 460 GB/s | 1792 GB/s |
| Price (approx) | $3,599 | $4,399 |
| LLMs it runs | 34 of 48 | 35 of 48 |
| Best model it runs | Gemma 4 31B · 15–22 tok/s | Gemma 4 31B · 50–75 tok/s |
The NVIDIA RTX 5090 (32GB) runs 1 more of the tracked models (35 vs 34). With the same nominal memory, more of it is usable for model weights on this architecture. The NVIDIA RTX 5090 (32GB) has more memory bandwidth (1792 GB/s), so it generates tokens faster at the same model and quant.